TO DETERMINE WHAT EFFECT INTRALIST SIMILARITY AND STIMULUS ISOLATION WOULD HAVE ON THE VON RESTORFF EFFECT, ISOLATED STIMULUS TERMS WERE PRESENTED IN RED AND NONISOLATED STIMULUS TERMS WERE PRESENTED IN BLACK. SUBJECTS WERE 60 FIRST GRADERS. THE PAIRED-ASSOCIATE ANTICIPATION PROCEDURE WAS USED. ACQUISITION TRIALS AND TRANSFER TESTS WERE GIVEN FOR EACH OF THREE LISTS OF TWO-LETTER WORDS CONSTRUCTED WITH AN ARTIFICIAL ALPHABET. ORTHOGONAL COMPARISONS INDICATED SIGNIFICANT DIFFERENCES FAVORING ISOLATED OVER NONISOLATED WORDS ON THE HIGH STIMULUS SIMILARITY LIST. FOR THE LOW STIMULUS SIMILARITY LIST, DIFFERENCES BETWEEN MEANS FOR NONISOLATED AND FORMERLY ISOLATED WORDS WERE NOT SIGNIFICANT. WHEN DISCRIMINATION ON THE BASIS OF LETTER FORM WAS EASIER, COLOR WAS A LESS IMPORTANT CUE. FOR THE THREE LISTS, IT APPEARED THAT DURING ACQUISITION, SUBJECTS TENDED TO USE LETTER FORM AS A CUE FOR RESPONSE EVEN FOR ISOLATED WORDS. THIS PAPER IS A REPRINT FROM "PROCEEDINGS OF THE 75TH ANNUAL CONVENTION OF THE AMERICAN PSYCHOLOGICAL ASSOCIATION, 1967," APA, WASHINGTON, D.C. (MC)